Steve Spencer
Offline
Upload & Sell: On
|
 Re: Strange thing happened on my way to the ideal kit... | |
ruthenium wrote:
Steve Spencer wrote:
ruthenium wrote:
Steve Spencer wrote:
ruthenium wrote:
Steve Spencer wrote:
ruthenium wrote:
OM-1 II has a better dynamic range vs. A9III at all equivalent ISO numbers (equivalent as in photographic equivalence).
Let me clarify this statement. What is not true is that at the same ISO (say ISO 800) that the OM-1 II has better dynamic range than the A9 III. The A9 III has about two-thirds of a stop better ISO when both cameras are set to ISO 800. What is true, however, is that for the same shutter speed (say 1/1000) that you can set the OM-1 II to two stops lower ISO, (say ISO 200 instead of ISO 800) and still get the same exposure from the two cameras as the 300 f/4 lens is two stops faster than the 100-400 GM with a 1.4X TC. This gives you more dynamic range for that shot that has a shutter speed of 1/1000 in either case.
This statement is confusing at best "the 300 f/4 lens is two stops faster than the 100-400 GM with a 1.4X TC. This gives you more dynamic range for that shot that has a shutter speed of 1/1000 in either case."
The mFT 300mm F4 and FF 400mm F5.6 with a ×1.4 TC are practically equivalent. Equivalent lenses allow the same amount of light to fall on the sensors of different sizes. When the SS is the same on these, the resulting images are expected to be equivalent. Equivalent images are theoretically expected to have the same or practically the same dynamic range. Thus, the advantage of the mFT system here is because the FF sensor of the A9III is different and has the dynamic range that is not exactly as good as that of the other FF Sony cameras.
No it isn't confusing. You just aren't understanding the ramifications of shining the light on a smaller area. The two lenses let in the same amount of light but the m4/3rds lens shines it on a smaller area so it is brighter and you can use a lower ISO. The FF sensor takes that same amount of light and shines it on a bigger area so the light is less bright across the bigger area so you need a higher ISO. What is happening here is no different than when the same flashlight projects a narrow beam and a wider beam. It projects the same amount of light in either case, but when it projects a narrower beam that narrower beam is brighter than when it projects a wider beam. Same amount of light but different brightnesses and when we are talking sensor sizes that means different ISO values with the same shutter speed.
Now it is true that bigger sensors with the same tech have better dynamic range than smaller sensors with that same tech and that means with the same tech you gain no dynamic range when you switch to a lens that projects to a smaller area. It is also true that the OM-1 II and the A9 III don't have the same tech or really even close to it and one of the tradeoffs for the global shutter in the A9 III is reduced dynamic range and that reduced dynamic range is why the OM-1 II has more dynamic range with a lens that let's in the same amount of light. It still doesn't change that at the same shutter speed, you can use a lower ISO with the OM-1 II than the A9 III and that is what is allowing the great dynamic range. If you shoot both cameras at the same ISO value and don't turn the ISO down on the m4/3rds camera you will both be shooting at a faster shutter speed (if you hold exposure constant) on the OM-1 II and you will have less dynamic range.
When you explain the "the ramifications of shining the light on a smaller area" you simply re-phrase what I said earlier, quote, "The ISO on sensors of different size is only comparable when related by the crop factor squared." Indeed, as you correctly stated this in your own words, given the same amount of light, the exposure (that you refer to as "brightness") of an mFT sensor is four times the exposure of a FF sensor, hence the latter MUST crank the gain (ISO) four-fold.
Your next statement "it is true that bigger sensors with the same tech have better dynamic range than smaller sensors with that same tech" is meaningless if you refer to the photographic dynamic range, because the PDR is not constant for a sensor and changes with the amount of light. What you are saying is like "the stomach of a large dog has more food than the stomach of a small dog" - but this depends on the circumstances, and may or may not be true, isn't it? For example, a small sensor that received more light than a larger sensor should have a better PDR. When the amount of light that falls on two sensors of different sizes is the same (as in the case of photogrpahic equivalence) then as I said earlier, quote, "Equivalent images are theoretically expected to have the same or practically the same dynamic range". Therefore, your next statement is incorrect "at the same shutter speed, you can use a lower ISO with the OM-1 II than the A9 III and that is what is allowing the great dynamic range" if you mean four times lower ISO with the OM-1 II (that is, an equivalent ISO on the OM-1 II).
Regarding the final claim "If you shoot both cameras at the same ISO value ... you will both be shooting at a faster shutter speed (if you hold exposure constant) on the OM-1 II"
Exposure is "the amount of light per unit area reaching the surface of an electronic image sensor" (https://en.wikipedia.org/wiki/Exposure_(photography).
When the ISO is the same on an mFT and a FF camera, these are equally exposed, hence the FF sensor collects four times more light(!) than the mFT sensor. Assuming the two systems have, for example, a 300mm F4 (mFT) and 600mm F4 (FF), respectively, the shutter speed is going to be the same on both cameras. The resulting images are, naturally, not going to be equivalent. Note that the greater amount of light that falls on the FF sensor in this example is because of the larger iris of the 600 mm F4 lens and has nothing to do with the size of the sensor.
Ok, let's start with what is simple. The cameras and lenses we are talking about are the Sony A9 III and the 100-400 GM with a 1.4X TC and the OM-1 II with the 300 f/4. The Sony lens is an f/8 lens. The Olympus lens is an f/4 lens. We know that exposure is a function of aperture, shutter speed, and ISO. We also know that at f/4 lens lets in two stops more light than an f/8 lens. So, if we hold shutter speed constant and make the same exposure and shoot the Olympus lens at f/4 and the Sony lens at f/8, and let auto ISO determine the ISO we will be shooting the Olympus system at two stops lower ISO than the Sony system. Surely you are not challenging that point.
Now we have to understand the physics formula for the total amount of light hitting the sensor and that formula is the brightness of the light (i.e., aperture in photographic terms), by the duration of the light (i.e., shutter speed in photographic terms) by the area the light hits (i.e., the sensor size in photographic terms). So when we say these two lenses are equivalent in the amount of light they let in and we know that they are equivalent because the f/4 lens let's in two stops more light than the f/8 lens in aperture, but the FF sensor is four times (i.e., two stops) bigger in area than the m4/3rds sensor.
So, if you understand these basics then you will be able to see that my above statements are correct. If you shoot the Sony system at ISO 800 and the Olympus system at ISO 800 (i.e., hold ISO constant) and you have the same exposure then the Oly system with its two stops faster lens will be shot at four times the shutter speed as the Sony system.
You are not wrong to say, "When the ISO is the same on an mFT and a FF camera, these are equally exposed, hence the FF sensor collects four times more light(!) than the mFT sensor. Assuming the two systems have, for example, a 300mm F4 (mFT) and 600mm F4 (FF," but we are not talking about that situation. We are not comparing the Oly 300 f/4 to the Sony 600 f/4. We are comparing the Old 300 f/4 to the Sony 100-400 with the 1.4X TC (i.e., an f/8 lens) so if we shoot the Oly at f/4 and the Sony at f/8 they will not have the same exposure at the same ISO if they are shot at the same shutter speed. I am pretty sure you know that.
Please reread what I wrote above in light of these simple principles and you will see that my statements are not incorrect. You should also note that we do not disagree that in this case because the A9 III sacrifices dynamic range to implement its global shutter that when the Oly system and the Sony system both shoot in the same light you will get more dynamic range from the Oly system. What you aren't understanding, however, is that you will only get that extra dynamic range if you are shooting the Oly system at a lower ISO than you are shooting the Sony system (and you can do that easily if you shoot at the same shutter speed), but if you shoot the two systems at the same ISO which means four times the shutter speed for the Oly system (as you are using an f/4 lens instead of an f/8 lens) you will get more dynamic range with the Sony system.
I shall try to keep it simple and respond to your points:
1) "We know that exposure is a function of aperture, shutter speed, and ISO"
Exposure is a function of ONLY aperture and shutter speed. The ISO is a function of exposure.
Just in case, I copy-pasted the definition of exposure: "Exposure is "the amount of light per unit area reaching the surface of an electronic image sensor."
2) "at f/4 lens lets in two stops more light than an f/8 lens" is correct when shooting the same system, e.g. any FF camera. An mFT 300mm f/4 and a FF 600mm f/8 are EQUIVALENT. They have the same size of the iris 300/4 = 600/8 = 75mm. Thus, THE SAME amount of light passes through these lenses and falls on the sensors.
3) "if we hold shutter speed constant ... and shoot the Olympus lens at f/4 and the Sony lens at f/8, and let auto ISO determine the ISO we will be shooting the Olympus system at two stops lower ISO than the Sony system."
This is correct, BECAUSE THE TWO SYSTEMS ARE EQUIVALENT.
A full-frame and a cropped camera system are expected to produce equivalent (indistinguishable) images when
(a) These are used from the same point, at the same distance to the same target (scene)
(b) The shutter speed is the same
(c) The focal lengths and the aperture f/numbers are related by the crop factor R
Under these conditions, the auto ISO (FF) must be R2 x auto ISO (mFT), where R2 stands for R squared.
4) "formula for the total amount of light hitting the sensor ...is the brightness of the light (i.e., aperture in photographic terms), by the duration of the light (i.e., shutter speed in photographic terms) by the area the light hits (i.e., the sensor size in photographic terms)"
This is incorrect. The total amount of light that falls on a sensor is determined by the field of view, by the physical size of the lens' entrance pupil and the time the sensor is open to the light. When two lenses are equivalent (i.e., they have the same field of view and the same size of the entrance pupil, see the condition in 3(c)), then the total amount of light received by sensors of different sizes is the same. In other words the amount of light is EXCLUSIVELY controlled by the lens and by the shutter speed.
5) "If you shoot the Sony system at ISO 800 and the Olympus system at ISO 800 (i.e., hold ISO constant) and you have the same exposure then the Oly system with its two stops faster lens will be shot at four times the shutter speed as the Sony system."
The same ISO automatically means the same exposure, and vice versa (see the definition above).
When using equivalent lenses (mFT 300mm f/4 and FF 600mm f/8), the same exposure is possible ONLY when four times more light is allowed to reach the four times larger FF sensor. As the lenses are equivalent, the only way this can physically happen is when the FF sensor is open to the light for a time that is four times longer than the mFT sensor. Thus, you are correct here.
6) "if we shoot the Oly at f/4 and the Sony at f/8 they will not have the same exposure at the same ISO."
This is incorrect, as already stated above: the same ISO means the same exposure, by definition.
7) "if you shoot the two systems at the same ISO which means four times the shutter speed for the Oly system (as you are using an f/4 lens instead of an f/8 lens) you will get more dynamic range with the Sony system"
This is correct, for the simple reason that under the condition of the same ISO, the FF sensor captures four times more light vs. the mFT sensor.
The same applies to two FF cameras if they have the same lenses, one open to f/4 and the other closed to f/8, the former system is going to produce images with a considerably better dynamic range.
The dynamic range drops if a system is starved of light. Naturally, when we compare different systems, we must compare them under equivalent conditions, that is by allowing the same amount of light to reach the sensors. Between the mFT and FF systems, this is the case only and strictly when ISO(FF) = 4 ISO(mFT). In other words, comparing a FF system and a cropped-sensor system at the same ISO is kind of "cheating" by giving the former four times more light and then (falsely) concluding the the latter is not good because it displays more noise.
First, you are wrong that #4 is incorrect. It is a basic law of physics and you botch your response by bringing field of view into discussions of how much light hits the sensor. I think we all know that the same amount of light hits the sensor on one of our cameras when we use the same aperture and shutter speed whether we shoot a 10mm lens or a 1200mm lens. The field of view is irrelevant to the amount of light hitting the sensor. That you bring up field of view makes it clear that you are confusing amount of light and equivalence. They are not the same concepts.
Now let's go back to #3. #3 is correct not because the lenses are equivalent but because the amount of light hitting the sensor is equivalent. An example can hopefully make that clear. If I shoot a 10mm lens on a m4/3rds camera at f/4 and and 1200mm lens on a FF camera at f/8 with the same shutter speed and with the FF camera at two stops higher ISO they will have the same exposure. You can check that out. These lenses are of course not equivalent. They are in fact very different. The reason they will have the same exposure is that the formula in #4 is correct and the total amount of hitting each sensor is the same. That total amount of light is independent of field of view (i.e., focal length). So you are just wrong that the reason that #3 is correct is that the lenses are equivalent and hopefully this example can illustrate to you why you are wrong.
Now let's go back to #2. It is not because the lenses are equivalent in focal length that they let in the same amount of light. It is because the aperture by the image circle they project are equivalent. Once again this calculation is independent of field of view (i.e., focal length). A 10mm lens at f/4 that projects a 21.6mm image circle (i.e., m4/3rds) and a 1200mm f/8 lens that projects a 43.3mm image circle (i.e., FF) will shine the same amount of light on the sensor. Field of view or focal length has nothing to do with it.
Now let's go back to #1. We are simply using the word exposure here differently. You certainly know that if you shoot a shot with a 1/100 shutter speed, an f/4 aperture, and ISO 1600, and if it is exposed well when I look at the picture that if I turn the ISO down to ISO 100 it will affect the image. It will in fact make the image quite dark. That presentation of the image is what I mean by exposure. And my usage of exposure here is not uncommon. In fact, many guides to photography describe the exposure triangle as shutter speed, aperture, and ISO. So let's not fight about what the word exposure means. Let's just be clear about what it means and hopefully I am being clear here. What I mean is how bright versus dim the image appears before it is adjusted in post processing.
Now let's go back to #5. We agree on #5, which is good but do recognize we are using exposure in the different way that is described in the paragraph above. You are insisting on your usage, but I hope you will allow my use of the term as well.
Now let's return to #6. It is the same issue as you are not understanding my use of the term exposure.
Finally, let's return to #7. Yes the total amount of light hitting the sensor as described in #4 is central to the amount of dynamic range and at the same aperture as described in the formula in #4 larger sensors have more light hitting the sensor as the total amount of light is a function of sensor size (in area) by brightness of the light (photographically that can be measured by aperture) by length of time the light hits the sensor (photographically that can be measured by shutter speed). If aperture and shutter speed are held constant then the bigger area of a larger sensor means more light is hitting the sensor. That is the advantage larger sensors have. It is totally independent of field of view (i.e., focal length) of the lens.
Dynamic range, however, is not just about the total amount of light hitting the sensor. It is also a function of how effectively the sensor converts that light into a digital signal, and we know ISO matters for that. When we set our sensor to a higher ISO the sensor is more sensitive to the light hitting it, but it also is doesn't convert the light to signal as well as it does at lower ISO. That is why we get more noise at a higher ISO and why we get less dynamic range at a higher ISO. Sensors with different technology also don't convert the light hitting the sensor into a digital signal with the same effectiveness. Some sensor do the conversion better than others. The A9 III has a groundbreaking global shutter, but one of the drawbacks of that sensor is that it isn't quite as effective at converting the light that hits the sensor to digital signal as Sony's other sensors.
In #7 above you only want to make comparisons that hold the amount of light hitting the sensor constant, but I think that is limiting. We can do other comparisons, however, and I think it is useful to do so. I don't think it is cheating for example to compare the dynamic range between two camera using the same aperture, shutter speed, and ISO. When you do so, and one has a larger sensor, yes the camera with the larger sensor will have greater dynamic range if the sensors are equally capable of converting the light to a digital signal. If you only compare when the smaller sensor is at a lower ISO you are intentionally handicapping the larger sensor in a way that you don't have to do (you can turn the ISO down on the FF camera after all and as long as you don't hit too low of a shutter speed that is a fine strategy). So I reject your premise that if you are comparing lenses at the same aperture and shutter speed you shouldn't compare them at the same ISO. I think that is an important and useful comparison.
|
